Strategies of Desecuritization · Collection Strategies of Desecuritization By Edin Fako Submitted to Central European University Department of International Relations and European

CE

UeT

DC

olle

ctio

n

Strategies of Desecuritization

By Edin Fako

Submitted to Central European University

Department of International Relations and European Studies

In partial fulfillment of the requirements for the degree of Master of Arts

Supervisor: Professor Paul Roe

Word Count:15,463

Budapest, Hungary

2012

CE

UeT

DC

olle

ctio

n

i

Abstract

An important school within the field of security studies, the Copenhagen School takes a discursive approach to understanding how security is constructed. Their theory of securitization aims at explaining how issues become securitized, or taken out of the sphere of normal, deliberative politics, into the realm of emergency politics and extraordinary measures. Seeing security as doing more harm than good, the Copenhagen School prefers desecuritization, the lack of a language of emergency measures or existential threats. Theories of desecuritization have conceptualized strategies for desecuritizing migrant identity in Europe and national minorities in Central and Eastern Europe. This paper looks to analyze and evaluate these desecuritization strategies for new cases: deeply divided, post-conflict societies. The argument made here is that existing desecuritization strategies cannot be applied outside of a liberal democratic context. As such, a new strategy is provided: structural desecuritization through the implementation of power sharing mechanisms. By introducing the literature on power sharing as a guide to desecuritizing ethnic identity in post-conflict states, this paper aims at reconceptualizing existing strategies, pointing out the assumptions that inhibit their broader salience. The case of post-Dayton Bosnia is offered as an example of how structure and institutions can work to desecuritize ethnic tensions, that is, to bring ethnic relations back into the sphere of normal politics.

CE

UeT

DC

olle

ctio

n

ii

Acknowledgements Thank you to my friends on both sides of the Atlantic who supported me in writing this thesis. I am

also thankful to my parents, who have worked tirelessly for me to be where I am today. I would be

nowhere without them. Hvala, Mama i Babo. I could not have done this without the support of the

entire CEU staff and faculty, and to John Harbord in particular who read countless, very rough

drafts. Finally, thank you to m supervisor Paul Roe, who above all made this a thoroughly enjoyable

process.

CE

UeT

DC

olle

ctio

n

iii

Contents

Introduction ........................................................................................................... 1

1. Security, Securitization, and Desecuritization .................................................... 5

1.1 Security and securitization ............................................................................ 5

1.2 Desecuritization ............................................................................................ 9

1.2.1 Desecuritizing migrants in Europe ......................................................... 12

1.2.2 Desecuritizing national minorities ......................................................... 15

2. Managing ethnic divisions: power sharing as a desecuritization strategy in

deeply divided states ........................................................................................... 19

2.1 Deeply divided, post-conflict states ............................................................ 20

2.2 Limits of existing strategies for desecuritizing ethnic identity in deeply

divided, post-conflict states .............................................................................. 22

2.3 Institutional solutions for securitized identities: the case for power sharing

......................................................................................................................... 26

2.4 Power-sharing as desecuritization .............................................................. 29

2.5 What kind of politics? ................................................................................. 33

3. Institutional desecuritization in post-Dayton Bosnia ........................................ 37

3.1 From insecurity to security: the success of the Dayton Accords .................. 38

3.2 Consociational democracy in Bosnia ........................................................... 39

3.3 Institutional desecuritization in times of crisis ............................................ 41

Conclusions .......................................................................................................... 48

Works Cited ......................................................................................................... 50

CE

UeT

DC

olle

ctio

n

1

Introduction “Those who sacrifice liberty for security deserve neither.” – Benjamin Franklin

In the decades following the end of the Cold War, the field of security studies has been

inundated with new ways of thinking about international security. Dominant paradigms have been

challenged by academics unsatisfied with existing concepts, looking to explain security in a

transformed world. Primarily, they sought to move security studies beyond theories that recognized

only military threats as challenges to state security. One leading approach to conceptualizing security

is that of the Copenhagen School (CS) and their theory of securitization.

Barry Buzan and Ole Waever have written the bulk of the work of the CS, and like many

other recent schools of security studies, they aim at widening and deepening the concept of security

to accommodate it to a new, post-Cold War global political order. Securitization theory radically

breaks from traditional realist and neorealist principles in that it adopts a social constructivist to

understanding security. Unlike these earlier traditions, securitization theory conceptualizes security as

discursively established, dismissing outright the notion of objective threats. The CS also breaks from

the realist and neorealist traditions in introducing the concept of “society” alongside the state as an

object that can be threatened and is worthy of analysis.

Securitization theory aims at understanding how issues become securitized, focusing on the

role of speech in the framing of threats. In this framework, the role of the security analyst is not to

discover the objective reality of threats “out there,” but rather to understand how security dynamics

are discursively established. In understanding threats as subjectively constructed, the CS disavows

the analyst’s capacity to show what is and is not a threat; but this does not mean they see all

securitizations as equally legitimate. Instead, they maintain that by seeing all threats as subjective,

they restrict actors’ ability to legitimize harmful policies based on the threatening nature of an issue.

CE

UeT

DC

olle

ctio

n

2

For the CS, securitization – the elevation of an issue to that of threat – means removing an

issue from the realm of normal politics and placing it into the realm of extraordinary measures.

Because they conceptualize threats as existentially threatening, securitizing an issue allows actors to

rise above the norms that would normally bind them in order to eliminate the existential threat.

Consequently, the CS view of securitization is generally a negative one, as it allows actors to justify

all sorts of policies in order to deal with a discursively constructed threat. They argue that the aim

should be desecuritization, moving issues “out of emergency mode and into the normal bargaining

process of the political sphere.”1 Unfortunately, despite a preference for it, the CS leaves the concept

of desecuritization largely un-theorized.

Most of the existing literature on desecuritization has dealt with the securitization of

immigration in Europe with the purpose of developing desecuritization strategies that can reverse

this trend. Branching off from this, some scholars have proposed that the particular strategies for

removing immigration policy from the security sphere may not be possible or desirable in the case

of national minorities in Central and Eastern Europe. Desecuritization strategies outside of these

two contexts have been largely un-explored.

Securitization of migrants in Europe has led to a robust debate on the merits of approaching

immigration policy from a security perspective. This literature has explored the reasons for

securitizing migration policy, as well as strategizing ways to desecuritize this issue. Primarily, these

strategies have stressed the need to re-assert democratic principles in the face of un-democratic

policies that come with the securitization of migrants. Another strand of literature has dealt with

desecuritization of national minorities in Central and Eastern Europe, offering strategies that do not

require minorities to reject their distinct identities in favor of civic, non-ethnic or other “more

acceptable” ones. This latter scholarship offers a particular understanding of securitization that

1 Buzan, Barry, Ole Waever, and Jaap de Wilde. Security: a new framework for analysis. Boulder: Lynne Rienner, 1998, 4.

CE

UeT

DC

olle

ctio

n

3

might be helpful in theorizing desecuritization strategies for as of yet unexplored cases: deeply

divided, post conflict states.

In countries throughout the globe, mass violence and civil war has left states divided along

ethnic lines. From Nigeria to India to Bosnia and Northern Ireland, ethnic identity is framed in the

language of security; but for the CS and other scholars working on desecuritization, this is

unmapped territory. This is exceptionally problematic if one considers that these are places where

representing otherness as existentially threatening has resulted in the deaths of millions. Focusing on

the specific situation in Bosnia, I hope to theorize possible desecuritization strategies for deeply

divided, post-conflict states, simultaneously demonstrating the limitations of broadly applying the

same theory to different cases. The purpose is not to create an overarching theory of

desecuritization – a perhaps unnecessary and undesirable endeavor – but rather to begin a

conversation about desecuritization in uncharted waters.

Bosnia is an important case for two reasons. First, as one of the earliest and largest nation-

building efforts by the international community in the 20th century, it has been viewed as a model for

structuring other post-conflict ethnically divided states. The lessons learned in Bosnia have been

applied to Iraq, Afghanistan and Kosovo among others, and will undoubtedly continue to shape

how the world approaches similarly divided states. Secondly, as a rigidly designed consociational

democracy, Bosnia demonstrates the possibilities of institutional solutions to securitized identity, as

well as the limits.

To say that institutions have desecuritized ethnic identity does not mean that liberal

democracy has prevailed or that identity is not spoken about in a language of security. As will be

argued, desecuritization does not have to mean that there is no language of security, but rather that

security issues are discussed within the realm of normal, deliberative politics. Such an understanding

of desecuritization has been largely overlooked in the literature, which usually sees desecuritization

CE

UeT

DC

olle

ctio

n

4

as bringing an issue into a realm of a-security, where it is not even spoken about in a language of

security. In Bosnia, ethnic identity is still heavily politicized and often spoken about in a language of

security, but this is all within the established political framework within the realm of normal politics. This

is not an ideal situation, but desecuritization does not have to mean restoration of ideal liberal

democratic politics, only normal politics; for Bosnia, nationalist politics is normal politics.

The argument made here progresses in three parts. The first chapter is an overview of the

CS approach to security and securitization/desecuritization, as well as important critiques that

challenge some assumptions, methods and assertions made. I then look at some of the strategies that

have been suggested for desecuritizing identity, specifically migrant and minority identity. This will

naturally lead into the theoretical and practical possibilities for desecuritization in divided states, the

bulk of chapter two. An important part of this chapter is an examination of theories of power-

sharing in divided states, focusing on consociational theory in particular. By understanding how

these normative theories have conceptualized security and identity, I hope to ground the theoretical

proposals made here in solid empirics.

Lastly, the third chapter will be an analysis of how power-sharing institutions have

desecuritized identity in post-war Bosnia while simultaneously hindering further democratization.

This case is not meant illustrate one way or another the validity of the CS approach to security, but

rather to elucidate the implications of this sort of approach to security and the theoretical limits of

desecuritization in the existing literature. Ultimately, this paper seeks to engage with the debate on

security and power-sharing to expand the conversation on desecuritization. The proposition I make

is that the existing literature is largely unsuited for theorizing desecuritization in deeply divided, post-

conflict states, warranting an investigation of possible institutional solutions offered by scholars

outside of security studies.

CE

UeT

DC

olle

ctio

n

5

1. Security, Securitization, and Desecuritization Browsing through literature in international relations, one would quickly grasp a reoccurring

premise: security is a fundamentally contested concept. This has become more evident in recent decades

with the proliferation of postmodernist, constructivist, postcolonialist, feminist, and other “critical”

studies of security. Classifying such epistemologically and ontologically different approaches

together in one excessively broad category is analytically unhelpful, but it does uncover a common

thread. All these different ways of thinking about security have arisen out of a profound

dissatisfaction with the realist and neorealist schools that dominated international relations for much

of the 20th century.

Defined in this way, the CS can also be characterized as critical security studies, but this is

not an entirely accurate label. What then is the CS understanding of security? Before addressing

desecuritization strategies, it is crucial to clarify the CS’s particular take on security and to introduce

their theory of securitization. Understanding securitization is fundamental in developing and

building upon existing desecuritization strategies.

1.1 Security and securitization

The CS stands somewhere in between two vastly different perceptions of security, having

one foot in traditional realism/neorealism and the other in a tradition of peace studies and social

constructivism.2 Though securitization theory proceeds from a social constructivist understanding of

threats, the CS methodology is fairly objectivist, as they choose methodological collectivism over

analysis at the individual level. Additionally, unlike some other “non-traditional” schools of security

studies – critical and feminist security studies, for example, – they refrain from establishing an

2 McSweeney, Bill. "Identity and Security: Buzan and the Copenhagen School." Review of International Studies 22, no. 1 (1996): 81-93.

CE

UeT

DC

olle

ctio

n

6

emancipatory ideal. In other words, the CS does not define what security is, but rather chooses to

focus on understanding how “security” is constructed and what speaking “security” does.

How is security constructed? The CS view is that security is established through discourse, or

more specifically, through a speech act by a securitizing actor. It is the act of naming an existential

threat that legitimizes actors to take extraordinary measures and to break the rules that normally

bind them.3 Theoretically, any actor can speak on behalf of a referent in a securitizing move;

practically, however, securitizations are carried out by the traditional elites.4 Furthermore,

securitizations may fail if, for example, the securitizing agent fails to convince the relevant audience

that an issue is existentially threatening, or if the referent is deemed to be not worthy of being saved.

Security rests “among the subjects,” [emphasis in original] suggesting that securitization is ultimately a

negotiated process.5

A securitizing move is made more likely by the presence of felicitous facilitating conditions

in three categories:

1) the demand internal to the speech act of following the grammar of security, 2) the social conditions regarding the position of authority for the securitizing actor – that is, the relationship between speaker and audience and thereby the likelihood of the audience accepting the claims made in a securitizing attempt, and 3) features of the alleged threat that either facilitate or impede securitization

[emphasis add].6

The emphasis on authority is significant in that it posits that securitization can only be carried out by

coercive and persuasive power, institutional or other. While the CS admits that power relations

between subjects inevitably play a role in a securitizing move, they claim that such power is “never

absolute,” emphasizing that no one is theoretically excluded from challenging a securitization.7

3 Buzan et al., Security: a new framework for analysis, 24-31. 4 Buzan, Barry, and Lene Hansen. The evolution of international security studies. Cambridge: Cambridge University Press, 2009, 214. 5 Buzan et al., Security: a new framework for analysis, 31. 6 Ibid., 33. 7 Ibid., 31.

CE

UeT

DC

olle

ctio

n

7

Nevertheless, they do not elaborate on how security could be vocalized or challenged by those

without power or authority

What sort of “extraordinary measures” and “emergency politics” does securitization

legitimize? The CS does not elaborate much, arguing that a securitization reveals itself once a

relevant audience has accepted that an issue must be dealt with outside of “rules that would

otherwise have to be obeyed.” In other words, securitizations have distinctive consequence,

depending on what constitutes “normal” politics there. It is not enough to have only extraordinary

measures or only the presence of existential threats, but the combination of “existential threats that

legitimize the breaking of rules.” 8 What may be extraordinary measures in one place could be

perfectly normal in another. I will return to this topic later when discussing what sort of politics

desecuritization seeks to (re)establish.

A number of scholars have criticized the CS’s emphasis on the discursive element in

securitization theory. What most of these critiques have in common is that they view the emphasis

on speech as problematic in cases where the ability to speak is constricted, or where securitization

occurs without a speech act. Exploring this matter comprehensively is outside the scope of this paper,

but some arguments are worth noting as they offer important insight into developing

desecuritization.

Writing on the securitization of migration in the US and the EU, Didier Bigo criticizes the

CS’s privileging of securitizing actors’ institutional power to speak security. Noting that the portrayal

of immigrants as criminals and threats to the state persists in spite of extensive evidence to the

contrary, he argues that the framing of migration as a security issue is a result of the waning

influence and legitimacy of security professionals after the Cold War. By framing immigrants as

threats to the internal peace and homogeneity of the state, security professionals reiterate the

8 Ibid., 25.

CE

UeT

DC

olle

ctio

n

8

necessity of their privileged position as bearers of special knowledge and power.9 Bigo’s criticism of

the CS’s conceptualization of security is that it reaffirms security professionals’ and other traditional

elites’ positions as legitimate possessors of knowledge and the power to name threats. By describing

security as a process of persuasion and coercion between elites and audience, the CS fails to

challenge traditional structures of power that deny security to those who are not authorized to speak

security. Though the CS explicitly denies the objective nature of threats, they do so implicitly by

accepting the security professionals’ truths about security and the “framing of a different domain of

security beyond the political.”10

Another critique of the discursive element in securitization is offered by Lene Hansen, who

reconsiders the implications of security conceptualized as speech act. She argues that by defining

security as an illocutionary performance done by authorized actors, securitization is an inaccessible

apparatus for those who are unauthorized to speak security. Hansen’s analysis is of women in

Pakistan who are victims of rape and sexual violence, arguing that they are silenced in the CS’s

discursive understanding of security. They are unable to securitize against the threat of sexual

violence because vocalizing that they were the survivors of rape or sexual assault would expose them

to the practice of honor killings. She concludes that this exposes two “blind spots” in the CS

understanding of securitization, where 1) securitization is difficult or impossible “in situations where

the possibilities of speaking security are constrained” [emphasis added] and 2) the CS ensures that

gender is incapable of becoming a referent object.11

Claire Wilkinson also challenges the conceptualization of securitization as a speech act,

making the case that securitization can be achieved in other ways as well. She contends that the

9 Bigo, Dider. "Security and Immigration: Toward a Critique of the Governmentality of Unease." Alternatives: Global, Local, Political 27, no. 1 (2002): 63-92. 10 Ibid., 73. 11 Hansen, Lene. "The Little Mermaid's Silent Security Dilemma and the Absence of Gender in the Copenhagen School."Millennium - Journal of International Studies 29, no. 2 (2000): 285-306.

CE

UeT

DC

olle

ctio

n

9

Tulip Revolution in Kyrgyzstan in 2005 demonstrates how a securitization can be achieved without a

speech act but rather through physical acts like mass mobilization and protest. In the face of an

existential threat to all people of Kyrgyzstan, numerous different tribal, ethnic, and patronage groups

mobilized against the dictatorship of Askar Akaev, ultimately ousting him and introducing new

elections.12 Wilkinson’s biggest contention is with “the constitution of agency proposed by

securitization,” which she believes is a symptom of the Westphalian straitjacket that limits the CS’s

ability to theorize outside out a Euro-centric frame.13

Introducing critiques of securitization theory’s emphasis on the discursive quality of security

demonstrates that securitization-as-speech act can be problematic. Though it provides an important

re-thinking of traditional paradigms, this conceptualization can be limiting for those wishing to

mobilize the power of security for a particular issue, or in demonstrating a securitization that has

occurred without a speech act. Wilkinson’s reference to a Westphalian straitjacket is also interesting

in that it suggests securitization theory makes certain assumptions that hinder its explanatory power

outside of a specific context. Consequently, it is possible that desecuritization is restricted by certain

assumptions as well. Specifically, if conceptualizing securitization as speech act has its limits, it is

plausible that desecuritization might also be possible outside of a discursive construction. Before I turn

to this more directly, I provide a review of the relevant literature on desecuritization.

1.2 Desecuritization

Securitization establishes a logic by which authorized actors can escape the normal realm of

deliberative politics – whatever they may be – and embrace a state of exceptionalism of expedient

decision-making in order to deal with a perceived existential threat. Security is a realm outside of

politics and the usual laws and norms that bind actors’ decisions. Consequently, the CS has a

12 Wilkinson, Claire. "The Copenhagen School on Tour in Kyrgyzstan: Is Securitization Theory Useable Outside Europe?"Security Dialogue 38, no. 1 (2007): 5-25. 13 Ibid., 22.

CE

UeT

DC

olle

ctio

n

10

negative conception of security: the logic of security might be unavoidable, but it certainly should

not be stretched to engulf a wide range of issues. Their position on this last point is worth quoting

at length:

“Basically, security should be seen as negative, as a failure to deal with issues as normal politics. Ideally, politics should be able to unfold according to routine procedures without this extraordinary elevation of specific ‘threats’ to a prepolitical immediacy. In some cases securitization of issues is unavoidable, as when states are faced with an implacable or barbarian aggressor. Because of its prioritizing imperative, securitization also has tactical attractions – for example, as a way to obtain sufficient attention for environmental problems. But desecuritization is the optimal long-range option, since it means not to have issues phrased as ‘threats against which we have countermeasures’ but to move them out of this

threat-defense sequences and into the ordinary public sphere” [emphasis added].14 What more does the CS have to say about desecuritization strategies? Unfortunately, not

much. There is clearly a preference for “routine procedures” and “ordinary politics,” but as was

mentioned earlier, the subjective nature of “normal” politics means this is a deliberately unspecified

concept. The explicit suggestion is clearly that debate and contestation are preferable to emergency

politics, but it is unclear whether this means democracy. As will be argued later on, this vagueness in

what constitutes normal politics is very problematic when ethnic politics is normal politics and

identities have become institutionally securitized. Before I make this argument, however, it is

necessary to look over desecuritization literature more broadly.

Ole Waever theorizes three strategies of desecuritization: not speaking about an issue as a

threat at all, managing a securitization so that it does not spiral, and moving the securitized issue

back into normal politics.15 In subsequent literature, desecuritization has generally become associated

with this last strategy. Before turning to the third strategy, however, the first two deserve additional

scrutiny

14 Buzan et al., Security: a new framework for analysis, 29. 15 Waever, Ole. "The EU as a Security Actor: Reflections from a Pessimistic Constructivist on Post-Sovereign Security Orders." In International Relations Theory and the Politics of European Integration: Power, Security, and Community, Morten Kelstrup and Michael C. Williams, 250-294. London: Routledge, 2000, 253.

CE

UeT

DC

olle

ctio

n

11

In discussing the emergence of the European “non-war community,” Waever defines three

different phases in inter-European relations: insecurity (1940s and 1950s), security (1960s),

desecuritization (1970s to mid-1980s), and finally, re-securitization in the 1990s.16 Because he defines

the period of desecuritization in Europe as one of a-security – where “the very question of what

kind of security arrangement one relied on became absurd”17 – it is difficult to see how management

of a securitization (so does it does not “generate security dilemmas and other vicious spirals”)18 can

be seen as a desecuritization. This last point is crucial for conceptualizing desecuritization in deeply

divided, post-conflict states and I will return to it later. Not speaking about an issue as a security

threat in the first place is also an interesting desecuritization strategy, as it implies a lack of an initial

securitization. This point does not need further elaboration, suffice to say that it demonstrates

Waever’s – and by extension the CS’s – pessimism about the possibilities of desecuritizing

securitized issues, especially if they are issues regarding societal security.19

Having discussed the first two strategies of desecuritization, I move now to the literature

that generally views desecuritization through the lens of the third strategy: moving issues back into

the realm of normal politics. How can this be achieved? First, the literature suggests that there is not

one general strategy for this. Depending on the nature of the securitized issue, some strategies will

undoubtedly work better than others will. Second, desecuritization strategies aiming at the

restoration of normal politics do not seem to share the same idea about what constitutes normal

politics. Consequently, they do not follow the CS in viewing “normal” politics as subjective, but

rather posit normative ideas on what these politics should be.

16 Waever, Ole. "Insecurity, security, and asecurity in the West European non-war community." In Security Communities, Emanuel Adler and Michael Barnett, 69-118. Cambridge: Cambridge University Press, 1998, 69. 17 Ibid., 86. 18 Waever, Ole. "The EU as a Security Actor: Reflections from a Pessimistic Constructivist on Post-Sovereign Security Orders," 253. 19 Ibid., 254.

CE

UeT

DC

olle

ctio

n

12

Approaches that perceive desecuritization as the restoration of normal politics can be

grouped into four categories: deconstructivist, emancipatory, reconstructivist, and management. The

first two have been advanced as potential desecuritization strategies primarily in the case of

securitized immigration policy; the latter two have been suggested for desecuritizing national

minorities. This paper seeks to build on the last strategy, that of managing ethnic identities. Before

this can be done, however, an analysis of the other strategies is necessary to contextualize the

argument made here.

1.2.1 Desecuritizing migrants in Europe

Concerned with the logic of security and the introduction of a political realist friend/enemy

dichotomy that securitization brings, Jef Huysmans posits a deconstructivist strategy of

desecuritization. He argues that the securitization of migrants in Europe is a security drama that, like

all security dramas in international relations, recreates a Hobbesian narrative of a war of all against

all. In this securitization narrative, the securitized migrant is the Other that has breached the

harmonious inner sphere – the mythologized homogenous state – and in doing so, has introduced

the possibility of death. Here, death is not only physical death but also the death of the discursively

constructed native identity through contact with non-native culture and customs. Most importantly

for Huysmans, it is in this portrayal of the migrant as a threatening other that the threatened identity

is also constructed:

“in creating threats – disharmony – the units create also their identity. This means that units and their identities are never just given in a security story, but that they develop within the story by the definition of threats.”20

20 Huysmans, Jef. "Migrants as a Security Problem: Dangers of "Securitizing” Societal Issues." In Migration and European Integration. The Dynamics of Inclusion and Exclusion, Dietrich Thränhardt and Robert Miles, 53-72. London: Pinter, 1995, 57-58.

CE

UeT

DC

olle

ctio

n

13

For Huysmans, this securitization story introduces a mutually constitutive friend/enemy

dichotomy present in a Schmittian conceptualization of the political community. Furthermore, such

a view of the political community is but one choice, one chosen technique that seeks to answer the

fundamental question Carl Schmitt’s theory sought to address:

“how to formulate a positive concept of the political which allows value determination in the context of modern societies torn between a formal process of rationalization and an aesthetic, irrational, subjective process of value determination which cannot be a successful counter-force to societal

rationalization because of its subjectivism.”21

In place of Schmittian political realism, Huysmans offers three different strategies of

desecuritization, objectivist, constructivist, and deconstructivist, ultimately settling on the latter as

most appropriate.22

A deconstructivist desecuritization strategy for desecuritizing migrant identity comes down

to a fragmentation, breaking down the “unified cultural alien” category and replacing it with a

plethora of shifting identities. Thus, a migrant is not a migrant but “woman, black, worker, mother,

etc. – just like the natives are.”23 Such a fragmentation of identities does carry with it a problem,

however, in that if carried to its logical conclusion, it allows no identity. Huysmans recognizes this,

and suggests the possibility of a simultaneous identity construction process to go along with identity

fragmentation, only to ensure that no one identity becomes dominant. Contemplating how to

formulate a political community with values but without a Schmittian logic of exclusion, his

response is to define the political sphere “in terms of the complexity and plurality of daily human

practices.”24

21 Huysmans, Jef. "The Question of the Limit: Desecuritisation and the Aesthetics of Horror in Political Realism." Millennium - Journal of International Studies 27, no. 3 (1998): 588. 22 Huysmans, Jef. "Migrants as a Security Problem: Dangers of "Securitizing” Societal Issues,” 67-68. 23 Ibid., 67. 24 Huysmans, Jef. "The Question of the Limit: Desecuritisation and the Aesthetics of Horror in Political Realism,” 588.

CE

UeT

DC

olle

ctio

n

14

Also reflecting on strategies to desecuritize the migrant identity in the EU, Claudia Aradau

offers emancipation – inspired by post-Marxist philosophers Etienne Balibar and Jacques Ranciere –

as a possible strategy that can overcome the non-democratic politics present in securitization. She

argues that securitization has an implicit Schmittian understanding of politics, where “securitization

is not simply a speech act,” but an “enactment of exceptionalism in political life.”25 Exceptionality

can have terrible consequences for personal liberties and democracy, as numerous authors have

demonstrated.26, 27

In place of the exclusionary and exceptional politics of securitization, Aradau maintains that

emancipation can re-start democratic process and introduce a universalist logic of mutual

recognition. In line with what the CS might argue, she rejects the Welsh School’s approach that

seeks to significantly expand the logic of security, equating security with emancipation. Such an

approach cannot mean true emancipation because it perpetuates the same Schmittian logic: some

must be made insecure for others to be secure.28 What Aradau suggests is that in order to reverse the

exclusionary logic of security present securitization, there must be “a process of dis-identification, a

rupture from the assigned identity and a partaking of a universal principle.”29

Andreas Behnke has challenged the idea that dis-identification is the right approach, arguing

that it demands exclusion of the “identity of the subaltern, the ‘security have not’s,’ as different”

[emphasis in original].30 Influenced by the work of David Campbell on US foreign policy and the

construction of national identity, he believes that desecuritization theorized as emancipation cannot

25 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation." Journal of International Relations and Development 7, no. 4 (2004): 392. 26 Huysmans, Jef. "The European Union and the Securitization of Migration." Journal of Common Market Studies 38, no. 5 (2000): 751-777. 27 Ceyhan, Ayşe, and Anastassia Tsoukala. "The Securitization of Migration in Western Societies: Ambivalent Discourses and Policies." Alternatives: Global, Local, Political 27 (2002): 21-39. 28 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation,” 397-398. 29 Ibid., 402. 30 Behnke, Andreas. "No Way Out: Desecuritization, Emancipation and the Eternal Return of the Political — a Reply to Aradau."Journal of International Relations and Development 9, no. 1 (2006): 66.

CE

UeT

DC

olle

ctio

n

15

happen. In order to ensure ontological security, states must continually produce an exclusionary,

securitizing logic. Thus, “the price of emancipation…is the elimination of difference” because only

by sacrificing their distinct social identity can the threatening other be accepted into the state and

dominant community.31 What are the implications of this for to other securitizations? Turning

eastward, I now examine strategies of desecuritization offered for national minorities in CEE.

1.2.2 Desecuritizing national minorities

States in CEE have historically had a difficult time accommodating national minorities,

leading to repressive policies and a fear that giving more rights and autonomy will lead to

secession.32 Consequently, the presence of these minorities has been framed as a security issue.

Collective ethnic identities differ in fundamental ways from the migrant identity imposed on

immigrants coming into the EU and US in the latter half of the past century, requiring different

strategies of desecuritization. Two approaches have been theorized to how the identity of these

minorities might be desecuritized: a management approach and a reconstructivist approach.

Paul Roe has argued that there is an inherent “securityness” to the collective identites of

national minorities that makes a deconstructivist or emancipatory desecuritization strategy logically

impossible. A deconstruction of the national minority’s identity would signify the death of group

identity because when speaking about the security of society, the group’s distinct identity is what

allows the group to survive. Thus, while states’ survival pins on their sovereignty, a society’s survival

is based on its distinct identity.33 Similarly, an emancipatory logic that demands dis-identification

runs the same risk of stifling the group’s identity. In place of these, Roe recommends a management

strategy to desecuritize national minorities.

31 Ibid., 67. 32 Micgiel, John S. State and Nation Building in East Central Europe: Contemporary Perspectives. New York: Columbia University Press, 1996. 33 Roe, Paul. "Securitization and Minority Rights: Conditions of Desecuritization." Security Dialogue 35, no. 3 (2004): 279-294.

CE

UeT

DC

olle

ctio

n

16

Due to the inherent nature of collective ethnic identities like those of Central and East

European minorities, Roe argues that the possibilities for desecuritization are limited. Rejecting the

idea that desecuritization is impossible entirely, he does note that it might be difficult practically. His

answer to the problem of securitized minorities is “management,” or moderate securitization.

Ultimately, Roe believes that it is difficult to avoid speaking of minority rights in terms of security,

but management can allow for a normalization of relations. What this means practically is a federal

solution to institutionalize mechanisms for deliberation that would reduce the necessity for

emergency politics.34

One response to Roe’s argument about the inherent “securityness” of minority rights and

the necessity of managing identities comes from Matti Jutila, who argues that Roe is deterministically

“writing security” into minority rights. Jutila agrees with what he believes is a tautological statement

that once a group’s distinctiveness is destroyed, it cease to exist. However, he does not see this

inevitably leading to security dilemmas between majorities and minorities. Rather, he supposes that a

reconstruction of identity, changing how groups see one another, is a superior desecuritization

strategy that does not “write security” into minority rights.35 In response, Roe has written that his

and Jutila’s approaches are not very different: a reconstructivist approach must be preceded by

management strategies. Mechanisms and institutions must be put in place to accommodate different

groups’ security concerns; only once this has been done can a reconstruction occur.36

What can be gleaned from the insights on desecuritizing ethnic minorities in CEE? Both

authors recognize the uniqueness of ethnic minority identity and correctly reject deconstructivist and

emancipatory desecuritization strategies due to the necessity of dis-identification. What are some

important gaps and problems with both arguments? For one thing, Roe’s advocacy for a “moderate”

34 Ibid., 292-293. 35 Jutila, Matti. "Desecuritizing Minority Rights: Against Determinism." Security Dialogue 37, no. 2 (2006): 167-185. 36 Roe, Paul. "Reconstructing Identities or Managing Minorities? Desecuritizing Minority Rights: A Response to Jutila." Security Dialogue 37, no. 3 (2006): 425-438.

CE

UeT

DC

olle

ctio

n

17

securitization of ethnic identity is puzzling. If the logic of securitization follows Schmittian political

realist lines, as many claim it does, “moderate securitization” is an oxymoron: one is either friend or

enemy; there is no room for an intermediate position. Though this at first appears like a fair critique,

in the next chapter I will argue that a moderate securitization is indeed theoretically possible.

Jutila’s advocacy for reconstruction is also problematic in that in practical terms, it is difficult

to see how this reconstruction could happen. He accepts that securitization is essentially about

narratives and that some will inevitably be able to assert particular narratives more effectively than

others. If we assume, as the CS and most others do, that those who can speak security are generally

the traditional state and societal elites, there seems to be no reason why elites would choose to

reconstruct narratives when the securitization of national minorities bestows upon them power and

resources they would otherwise not possess. Ultimately, reconstruction may be a theoretically sound

strategy of desecuritization but it is practically difficult and maybe even impossible. As Roe correctly

argues, what is first needed is management of ethnic divisions.

These are but some of the important questions and strategies to be considered when

approaching desecuritization. Both Roe and Jutila offer important insight into a topic few have

researched: desecuritizing ethnic minorities. Roe’s insistence on the establishment of federal

institutions and mechanisms to manage groups’ security concerns is spot-on. Furthermore, it

implicitly recognizes that the primary method of managing ethnic tensions in practice has been the

creation of power-sharing structures such as federalism. Jutila’s insights are also useful in further

developing possibilities of desecuritization because they emphasize the necessity of shifting

discourse. Importantly, however, these arguments deal with national minorities; the dynamics of

desecuritization are different in deeply divided, post-conflict states.

In the next chapter, I turn directly to the topic of desecuritizing ethnic identity in divided

states. While similar to national minorities in CEE, these cases are different in that they are even

CE

UeT

DC

olle

ctio

n

18

more divided due to the presence of recent mass violence. The insights and critiques reviewed thus

far will help elucidate how desecuritization can be theoretically conceptualized and practically

executed in such cases. An argument will be made for the necessity of power-sharing institutions

that go beyond federalism, creating a specific kind of political community necessary for

desecuritization of ethnic identity.

CE

UeT

DC

olle

ctio

n

19

2. Managing ethnic divisions: power sharing as a desecuritization

strategy in deeply divided states

How can desecuritization of ethnic identities occur in deeply divided states? Why do we

need a separate desecuritization strategy for such cases? What are deeply divided states? These are

but some of the questions this chapter will try to answer. This inquiry stems from dissatisfaction

with the existing literature’s failure to address the security dynamics in “deeply divided, post-conflict

states,” a label I will explain shortly. Proceeding from the previous chapter’s discussion of

desecuritization strategies, this chapter further develops the idea of federalism-as-desecuritization or

“management of ethnic identities.” If managing ethnic divisions suggests possible institutional

solutions for securitized ethnic identities, then it might very well be time for security studies to look

elsewhere for insights into theories of desecuritization. The vast literature on international nation

building, ethnic conflict management and democratization in political science and peace and conflict

studies is a good place to start.

Building upon Roe’s case for federal solutions to securitized national minorities, this chapter

explores other institutional desecuritization strategies in ethnically divided states where federalism

may be unfeasible. Specifically, I advocate for the desirability of consociational arrangements in

states where violent inter-ethnic conflict has ruptured the existing political landscape. These cases

are similar in some ways to states in CEE where national minorities are framed in security discourse

but also demonstrate fundamental differences that warrant separate analysis. Understanding the

unique security dynamics of post-conflict states divided by ethnic divisions is essential if we seek to

avoid further violence and provide for the emergence of democratic politics.

The argument for consociational institutions develops in three ways. First, the parameters

for “deeply divided, post-conflict states” will be established. The purpose is not to develop a

comprehensive typology but rather to demonstrate why this variety of states is unique and why they

CE

UeT

DC

olle

ctio

n

20

require different desecuritization strategies. Second, there must be further analysis of the established

desecuritization strategies, explaining why exactly such strategies are not desirable in post-conflict

states. Finally, using a particular understanding of desecuritization – that is, what it entails and what

it aims for – I develop a strategy of institutional desecuritization for deeply divided, post-conflict

states.

2.1 Deeply divided, post-conflict states

Developing a theory of desecuritization for deeply divided states requires an initial

clarification: what are deeply divided states? All states contain societal cleavages based on class, race,

religion, ethnicity etc. that influence politics, so what makes a state deeply divided? Understanding this

seemingly arbitrary delineation is important to the argument made here because it is exactly the

nature of these divisions that warrants a separate desecuritization strategy. Furthermore, the added

qualification of “post-conflict” is a further narrowing of the sort of states that may require special

desecuritization strategies from the ones in the existing literature.

Deeply divided, post-conflict states are usually states split along ethno-religious lines.

Perhaps unsurprisingly, many are also contested states, where one or more groups challenge the very

existence of the state in the established form – Nigeria, Iraq, Pakistan and Kosovo are but a few

examples. Consequently, deeply divided states are home to one or more secessionist movements

seeking to establish full sovereignty for their group. With a few important exceptions, academics and

policy makers alike have worked to prevent secession, as it can never be a decisive solution in

contested states: partition usually only inverts minority/majority positions.37

What is the nature of ethno-religious divisions that plague these contested states? Avoiding a

comprehensive review of the literature on nationalism and ethnic identity, two important comments

37 Sisk, Timothy D. Power Sharing and International Mediation in Ethnic Conflicts. Washington DC: United States Institute of Peace, 1996.

CE

UeT

DC

olle

ctio

n

21

are necessary. First, this paper rejects a primordial or objectivist view of ethnic or national identity.

There is extensive evidence to suggest that these identities are modern, socially constructed, and

negotiated38; they are not timeless39 or passed down through bloodlines.40 Nevertheless, this is not to

dismiss nationalism or ethnicity entirely. For example, Clifford Geertz41 argues that it does not

matter whether or not a nation has a particular lineage or history, what matters is that the perception

of this connection has great significance for social interaction.42 This is particularly true in post-

conflict states where the violence has been framed in the language of inter-ethnic animosity. Such

violence tends to increase individuals’ affinity with their own group and intensify animosity for

others. Thus, when considering strategies of desecuritization after mass violence framed in ethnic

terms has occurred, dismissing ethnicity entirely – in favor of a common, civic identity, for example

– will probably not be an effective strategy.

Second, the labels “deeply divided” and “post-conflict” are designed to accommodate a wide

range of states that have largely been ignored by both the CS and other academics working with

securitization and desecuritization. The category is meant only to highlight the fact that the implicit

assumptions the CS and others make in regards to state structure may not be present outside of a

liberal democratic context.43 Consequently, these assumptions limit some desecuritization strategies

that have been offered. Returning to the literature on desecuritization discussed earlier, I now

examine more closely the inapplicability of these strategies for deeply divided states.

38 Brubaker, Rogers. Nationalism Reframed: Nationhood and the National Question in the New Europe. Cambridge: Cambridge University Press, 1996. 39 Hobsbawm, E J. Nations and Nationalism Since 1780: Programme, Myth, Reality. 2nd ed. Cambridge: Cambridge University Press, 1992. 40 Balibar, Étienne. "The Nation Form: History and Ideology." In Becoming National: A Reader., Geoff Eley and Ronald G. Suny, New York: Oxford University Press, 1996. 41 Geertz, Clifford. The Interpretation Of Cultures. New York: Basic Books, 1973. 42 The CS takes a similar approach: recognizing the socially constructed nature of ethnic groups, they nevertheless approach ethnic identity as something that is often “solidly sedimented,” sometimes becoming a referent object for securitizing actors, see Buzan, Barry, and Ole Waever. "Slippery? contradictory? sociologically untenable? The Copenhagen schools replies." Review of International Studies 23 (1997): 242-244. 43 for more on CS bias, see Wilkinson, Claire. "The Copenhagen School on Tour in Kyrgyzstan: Is Securitization Theory Useable Outside Europe?" 2007.

CE

UeT

DC

olle

ctio

n

22

2.2 Limits of existing strategies for desecuritizing ethnic identity in deeply divided, post-

conflict states

The existing theories of desecuritization primarily deal only with two specific cases of

securitization: securitized migrants in Europe44, 45 and the securitization of national minorities in

Central and Eastern Europe.46, 47 Within the CS itself, desecuritization has been written about

primarily as a state of a-security that developed in Western Europe and Scandinavia in the latter half

of the 20th century.48 An analysis of how this state of a-security came about is largely un-theorized by

the CS. Consequently, we are left with the four types of desecuritization discussed in chapter one:

deconstruction, emancipation, reconstruction, and management.

Roe offers a valuable account of why the deconstructivist approach to desecuritization,

useful for desecuritizing migrants, may not be appropriate for national minorities in CEE. It is not

necessary to repeat his argument entirely, suffice to say that unlike migrants, ethnic groups might be

much less willing to shed their distinct identities in favor of shifting, fragmented ones.49 In deeply

divided states, identities may be even more rigid than in states where conflict with other ethnic

groups has not recently occurred. In other words, a deconstructivist strategy for desecuritization,

while theoretically sound, would probably be practically impossible to pursue.

An emancipatory strategy runs into a similar problem in that it requires dis-identification

from a particular identity. According to Aradau, the key to desecuritizing migrant identity is to

enable the securitized migrant to invoke “existing principles already present in a democratic

44 Huysmans, Jef. "Migrants as a Security Problem: Dangers of "Securitizing” Societal Issues.” 45 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation.” 46 Roe, Paul. "Securitization and Minority Rights: Conditions of Desecuritization.” 47 Jutila, Matti. "Desecuritizing Minority Rights: Against Determinism." 48 Waever, Ole. "The EU as a Security Actor: Reflections from a Pessimistic Constructivist on Post-Sovereign Security Orders." 49 Roe, Paul. "Securitization and Minority Rights: Conditions of Desecuritization.”

CE

UeT

DC

olle

ctio

n

23

regime.”50 There are two main problems with this approach. First, as Behnke correctly points out,

any political community demanding dis-identification in favor of universal principles inevitably

demands homogeneity and assimilation – the price of emancipation. 51 His contention rests on the

conflict between emancipatory logic and ontological security. Although a thorough engagement with

ontological security is outside the scope of this paper, it does warrant a brief consideration.

Behnke’s critique of emancipatory logic is that it ignores the inherent necessity of difference

in constructing a stable political community. Catarina Kinnvall argues that religion and nationalism

provide especially powerful sources of ontological security because they are portrayed “as resting on

solid ground, as being true, thus creating a sense that the world really is what it appears to be.”52 In

other words, an ethnic identity provides a secure sense of self that is strengthened in times of

instability, such as that brought on by globalization – Kinnvall’s argument – or by violence and a

rupture with established routines that give people meaning.53 Ultimately, any desecuritization strategy

that necessitates a rupture with such a potentially powerful sense of stability like an ethnic identity

might very well be impossible to implement.

Secondly, in advocating for an emancipatory strategy of desecuritization, Aradau seeks not

simply the restoration of “normal” politics, but the creation of a particular kind of political

arrangement. Consequently, it is obvious that an emancipatory strategy can only work in liberal

democracies; states where democracy is not well established or where the political arrangements

have been shattered by war and violence cannot easily uphold universal principles. Furthermore, in

emphasizing universal principles and dis-identification, Aradau implicitly advocates for the creation

50 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation,” 401. 51 Behnke, Andreas. "No Way Out: Desecuritization, Emancipation and the Eternal Return of the Political — a Reply to Aradau," 67. 52 Kinnvall, Catarina. "Globalization and Religious Nationalism: Self, Identity, and the Search for Ontological Security." Political Psychology 25, no. 5 (2004): 763. 53 For more on routine and ontological security in international relations, see Mitzen, Jennifer. "Ontological Security in World Politics: State Identity and the Security Dilemma." European Journal of International Relations 12, no. 2 (2006): 341-370.

CE

UeT

DC

olle

ctio

n

24

of a majoritarian democracy. In divided states, this may not be possible, desirable, or even necessary

for successful desecuritization.

While it should be clear by now why a deconstructivist or an emancipatory strategy of

desecuritization cannot work for deeply divided, post-conflict states, what are the problems with a

reconstructivist or management approach to desecuritizing ethnic identity? Both are valid

theoretically, but management is more likely to succeed in deeply divided, post-conflict states. On

the other hand, even management, posited simply as federalism, might not be enough. What is

needed is a deeper, more complex institutional solution. Before exploring this, however, a little more

needs to be said on reconstruction and management.

Responding to Jutila’s critique of his management strategy, Roe argues that while applicable,

the reconstructivist strategy must be preceded by a managing of minority rights.54 Considering the case

of national minorities in CEE to be particularly entrenched in a securitized logic, he contends that a

reconstruction, while necessary and desirable, can only proceed once there has been a period of

accommodation for minorities. For this purpose, he proposes the establishment of “cultural and/or

political autonomy as part of a federal structure.”55 These critiques are even more valid in the case of

post-conflict states, where the likelihood of shifting narratives about the former enemy is extremely

unlikely. Thus, management of some sort must precede reconstruction as a desecuritization strategy.

But is political/cultural autonomy in the form of multination federalism the only right solution?

While managing ethnic divisions in dynamics of security (where the minority is discussed in

the language of security but within an open, deliberative, and contestable political arrangement) is

desirable, it is less clear why management should be limited to federalism. Roe defers to Will

Kymlicka on this issue, who argues that desecuritization of minority rights in CEE must be

54 Roe, Paul. "Reconstructing Identities or Managing Minorities? Desecuritizing Minority Rights: A Response to Jutila," 432. 55 Ibid., 433.

CE

UeT

DC

olle

ctio

n

25

accompanied by territorial autonomy and the majority group accepting the possibility of secession.

Drawing on the Western experience with multiculturalism, Kymlicka concludes that vibrant

democracy can only occur once states accept “the possibility (however slim) of a democratically

mandated secession” by the minority.56 But minorities hardly ever occupy only one entirely

homogenous territory, and though he notes that the rights and identity of the newly created minority

must be protected, Kymlicka does not address the fact that allowing one secession can lead to others

(as the example of Serb secessionism post Bosnian and Croatian independence demonstrates). Is it

necessary to open the can of worms that is secession through territorial autonomy?

Even if territorial autonomy is the best option, why would a majority accept a federal

solution? If the securitizing narrative is that the minority represents a threat to the territorial integrity

of the nation-state, would they not see a federal solution simply as the first step to secession?

Federalism has been a solution to ethnic divisions in places like Belgium and Canada, but one need

only look to the violent break-up of Yugoslavia or Russia’s wars in the Caucasus to see the limits of

federalism. In the specific case of CEE, however, it is unclear how a federal solution could

successfully desecuritize national minorities, as it seems to confirm the majority’s worry that the

minority is a threat. Consequently, federalism and political/cultural autonomy may not be enough to

desecuritize ethnic identities in CEE; the chances for success are even lower in deeply divided, post-

conflict states. Instead, there must be deeper, more complex institutional solutions to securitization

of ethnic identity. In particular, there should be incentives for power to be shared, rather than just

federalized.

56 Kymlicka, Will. "Justice and security in the accommodation of minority nationalism." In Ethnicity, Nationalism, and Minority Rights, Stephen May, Tariq Modood, and Judith Squires, 144-174. Cambridge: Cambridge University Press, 2004, 164.

CE

UeT

DC

olle

ctio

n

26

2.3 Institutional solutions for securitized identities: the case for power sharing

How can desecuritization be achieved institutionally? Roe’s argument on the desirability of

federalism for managing ethnic divisions in cases of securitization is a good starting point, but

federalism alone may not be enough. Moreover, while territorial federalism might be outwardly

rejected by the dominant group due to the increased likeness of secession it is thought to create,

power-sharing might be a more suitable option. Before arguing for the suitability of power-sharing

as a desecuritization strategy in ethnically divided states, I provide an overview of the concept itself

and the larger debate on institutionally managing divided societies.

Power-sharing is a concept developed in political science that describes institutional and

electoral structures designed to be more inclusive of deep cleavages in a state where majoritarian

democracy is undesirable. No two forms of power-sharing are the same, and may even differ within

the state depending on what level of governance one observes. Consociationalism, theorized by

Arend Lijphart in his pioneering work The Politics of Accommodation: Pluralism and Democracy in the

Netherlands, is perhaps the most well-known version. According to Lijphart, a consociational

democracy contains four main features:

1) Grand coalitions of segmental elites that work together to stabilize the system 2) The possession of mutual vetoes by all factions on vital national interests 3) Segmented autonomy for all groups 4) Proportional representation in the electoral system57

A brief discussion of these four elements is needed. Lijphart warns that no one version of

any of these elements should be applied across the board, but instead, local conditions should shape

what kind of consociationalism a state adopts. Grand coalitions can be comprised of all relevant

segments of societies, or they may consist of the largest segments. What is crucial is that as many

57 Lijphart, Arend. The Politics of Accommodation: Pluralism and Democracy in the Netherlands. Berkeley: University of California Press, 1968.

CE

UeT

DC

olle

ctio

n

27

segments of society be included as is possible.58 Some states will undoubtedly be more inclusive than

others: Bosnia is divided between the three biggest groups, to the exclusion of other much smaller

minority groups, while Israel’s arrangement excludes Palestinians entirely.59

Furthermore, the use of “segmented autonomy” instead of simply territorial autonomy is a

nuanced solution to what can often be a complex problem. As discussed earlier, territorial autonomy

is often equated with impending secession in cases of securitized minorities, making it an

improbable desecuritization strategy. Autonomy can take many other forms, however, and while it

might be combined with territorial autonomy in some ways, it is important not to preclude other

considerations for autonomy when considering ethnic divisions.60

Lijphart conceptualizes consociationalism not as a recipe for how divided states should be

structured but rather as a guiding principle on how to make them more democratic. The problem

consociationalism seeks to address is the tyranny of the majority. Concerned with the possibility a

tyranny of the majority, Lijphart’s contends that while a majoritarian democracy may be suitable for

states that are largely homogenous, it is less suitable in states with deep societal cleavages. His

argument is not against majoritarian democracy but rather for the implementation of consociational

principles where they might help make the political landscape more inclusive. More importantly, the

choice may sometimes be for consociational democracy or no democracy at all.61

What is the nature of the divisions that consociationalism is supposed to manage? In his

early work, Lijphart studied the structures of power-sharing in his native Netherlands, “discovering”

consociationalism as a natural bargaining process that comes about in divided states. The divisions

in the Netherlands were ideological and religious, while in Belgium and Switzerland the divisions are

58 Lijphart, Arend. Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice. London: Routledge, 2008, 29. 59 O'Leary, Brendan. "Debating Consociational Politics: Normative and Explanatory Arguments." In From Power Sharing to Democracy Post-Conflict Institutions in Ethnically Divided Societies, Sid Noel, Toronto: McGill-Queens University Press, 2005, 14. 60 Lijphart, Arend. Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice, 2008. 61 O'Leary, Brendan. "Debating Consociational Politics: Normative and Explanatory Arguments," 6.

CE

UeT

DC

olle

ctio

n

28

national and linguistic. In Lebanon, the deep societal cleavages are primarily religious, while in

Bosnia they are ethnic. Thus, a state may be divided by any sort of cleavages, though religious,

linguistic and ethnic tend to be the most common.62

The foundational principle of consociationalism is the idea that democracy in deeply divided

states is better served by power-sharing rather than majoritarian principles. In other words, deep

cleavages along ethnic, religious, or other lines should be eased by recognizing these divisions and

incorporating them into the political sphere. Consociationalists are political realists who believe that

particular collective identities – ethnic, linguistic, national, or religious – are not fixed or primordial

but can certainly become fairly durable.63 Instead of wishing these differences away,

consociationalists believe it is preferable to work with the divisions and ensure that political

arrangements facilitate inclusion of all groups.

Consequently, one of the major critiques launched at consociationalists is that they reinforce

the same divisions they intend to alleviate. Integrationists like Donald L. Horowitz argue that

consociations fail at managing societal divisions because they do not allow for crosscutting

allegiances on civic or other less divisive principles.64 Power-sharing arrangements usually mean that

group leaders have little incentive to seek support outside of their own faction, possibly resulting in

nationalist outbidding and populist rhetoric.65 Instead, integrationists advocate for political

arrangements that minimize social divisions and increase incentives for inter-communal cooperation.

Like consociationalists, integrationists see ethnic, religious, and similar divisions as a social reality;

they are, however, more optimistic about the possibility of transcending these divisions.

An additional critique of consociationalism is that it is undemocratic: according to Lijphart,

the source of cooperation in a consociational democracy is the rule of a cartel of elites working

62 Lijphart, Arend. Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice, 2008. 63 O'Leary, Brendan. "Debating Consociational Politics: Normative and Explanatory Arguments," 8. 64 Horowitz, Donald L. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985. 65 Norris, Pippa. Driving Democracy: Do Power-Sharing Institutions Work? Cambridge: Cambridge University Press, 2008, 28.

CE

UeT

DC

olle

ctio

n

29

together to stabilize the state. Subsequently, many have disputed the democratic credentials of a

theory that necessitates backroom dealings by ethnic entrepreneurs to bring order and stability. In

response, consociationalists have argued that nothing in their theory precludes competition and

change of power within the different camps, only that political space be reserved for each segment of

society. Furthermore, nothing in consociational theory makes the dissolution of societal cleavages

impossible.66 This last point deserves further scrutiny, and I return to it later.

Integrationists and consociationalists approach the problem of divided states in a similar way

but disagree on how to implement change. Both schools see societal cleavages as a problem that can

be managed by creating incentives for cooperation; what they disagree on is the form of the

institutions that are needed. Though their view of the nature of the divisions they study may be

constructivist, this point is generally irrelevant, as their research methodology is entirely objectivist

and positivist. In fact, the resemblance between the CS understanding of identity and that of

consociationalists is striking: neither denies the socially constructed nature of ethnicity but both

proceed as if ethnic groups were social facts.

Can the debate between consociationalists and integrationists help in developing strategies of

desecuritizing ethnic identity? Undoubtedly, scholars working with desecuritization and those

theorizing institutional solutions for divided states have different epistemological and ontological

foundations, but this does not mean that they cannot learn from one another. If a federal solution

for desecuritization may not be acceptable, might a more complex power-sharing structure be

better?

2.4 Power-sharing as desecuritization

How might instituting power-sharing arrangements for divided states be conceptualized as a

strategy for desecuritization? A few stipulations are required. First, like desecuritization, power-

66 O'Leary, Brendan. "Debating Consociational Politics: Normative and Explanatory Arguments," 6-12.

CE

UeT

DC

olle

ctio

n

30

sharing should not be viewed as one comprehensive approach but rather as a procedural method for

restoring normalcy and deliberative processes. Not all issues can desecuritized in the same way and

not all states can have the same power-sharing institutions. What works for some may not work for

others. What must be the same is a preference for politics over violence, inclusion over exclusion,

and deliberative over emergency measures. The kind of politics is also important.

Second, while the discussion on federal solutions is specific to the desecuritization of ethnic

minorities in Central and Eastern Europe, power-sharing might be a desecuritization strategy in

many deeply divided, post-conflict states. The aim is not to create a comprehensive theory, however,

but rather to expand the institutional argument into another paradigm. Obviously, it would be

crucial to discern whether a securitization has in fact occurred. Answering positively would then

warrant an investigation into what is the nature of the securitization its effects.

Finally, in states where societal cleavages are deep enough to the point where self-identifying

groups see each other as security threats, institutional power-sharing can be an appropriate strategy

of desecuritization. Here I am equating deeply divided states with states that frame certain ethnic or

national in terms of security. In other words, a successful securitization can create a deeply divided

state; but a deeply divided state does not necessarily indicate a successful securitization. This point

does not need extensive elaboration; suffice to say, it is safe to assume that deeply divided states can

also be sites of securitized ethnic identities. Once it can be concluded that a divided state is also a

case of securitized identities, we can proceed with theorizing how power-sharing can be a useful

desecuritization strategy.

In order to understand how power-sharing could be seen as a valuable desecuritization

strategy, it is important to return to Waever’s differentiation between different states of security. Roe

correctly points out that while insecurity and a-security are fairly well understood, an often-

overlooked state is that of “security,” in which the language of security is present but it does not lead

CE

UeT

DC

olle

ctio

n

31

to the legitimation of extraordinary measures. He cites the debate and eventual banning of the

headscarf in France as an example of this state: French identity was clearly understood to be under

threat, but this did not lead to a violent spiraling or legitimation of emergency measures.67 Therefore,

the presence of a language of security does not reveal a securitization: as the CS write,

Securitization is not fulfilled only by breaking rules (which can take many forms) nor solely by existential threats (which can lead to nothing) but by cases of existential threats that legitimize the

breaking of rules.68

Based on this understanding of securitization, Roe goes on to argue that the management strategy of

desecuritizing national minorities in CEE might still leave the issue of re-securitization on the table,

but this is the best we can hope for in this situation. Without managing first, however, there can be

no reconstruction.

But what sort of desecuritization strategy is management? How can it be understood as part

of the established CS models of desecuritization? This answer requires two additional commentaries.

First, desecuritization should be understood as a process, one that must be continually reproduced.

Second, the aim of desecuritization does not have to be a-security, as Aradau69 and Huysmans70 have

understood it but can also aim for security.71 Returning to the conditions for securitization will allow

for a better understanding of how desecuritization can be viewed as a process.

Facilitating conditions are “the conditions under which the speech act works” and essentially

determine how likely a securitizing move is to succeed. Internal facilitating conditions are those that

demand the securitizing actor follows the established rules and procedures of speech act, while the

external conditions stipulate who is more likely to succeed in securitizing and what sort of things are

67 Roe, Paul. "Reconstructing Identities or Managing Minorities? Desecuritizing Minority Rights: A Response to Jutila." 68 Buzan et al., Security: a new framework for analysis, 25. 69 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation.” 70 Huysmans, Jef. "Migrants as a Security Problem: Dangers of "Securitizing” Societal Issues.” 71 Roe, Paul. "Reconstructing Identities or Managing Minorities? Desecuritizing Minority Rights: A Response to Jutila."

CE

UeT

DC

olle

ctio

n

32

more likely to be accepted as threatening.72 Thus, though the CS argues that the discursive nature of

security combined with the subjective character of threats means that any issue can be securitized –

widening the security agenda – in actuality, facilitating conditions limit what can convincingly be

argued as a security threat.

Furthermore, facilitating conditions – like the authority or social condition of the securitizer

or “features of the alleged threat” – are inherently variable and shifting. Consequently, a securitizing

actor must correctly interpret all these fluctuating conditions and intervene at exactly the right

moment in order to have the speech act accepted. If a speech act leads to a securitization, then it

would seem that the facilitating conditions that were meant to prevent such a move failed; such an

understanding reduces securitization to a moment.73 Increasingly, however, this conceptualization has

come under attack74, 75 as inaccurate in explaining securitization. Consequently, desecuritization

should not be understood as a moment either. Rather, it is a process that alters the facilitating

conditions for a speech act to prevent and hinder the logic of securitizations.

Understanding facilitating conditions is crucial for conceptualizing the implementation of

power-sharing arrangements as a desecuritization strategy. By limiting the ability of a securitizing

actor to persuade an audience about the threatening nature of an issue, threats can be desecuritized.

In the case of deeply divided societies, implementing power-sharing structures guarantees a voice to

all groups, ensuring that one dominant narrative will be much harder to establish by a securitizing

actor. If securitization is the elevation of threats above the realm of normal deliberative politics,

institutionally mandating that all groups be given the same authority to counter securitizing moves

against them could be a very effective desecuritization strategy. In other words, an institutional

72 Buzan et al., Security: a new framework for analysis, 32. 73 McDonald, Matt. "Securitization and the Construction of Security." European Journal of International Relations 14, no. 4 (2008): 575-577. 74 Guzzini, Stefano. "Securitization as a causal mechanism." Security Dialogue 42, no. 4-5 (2011): 329-341. 75 Huysmans, Jef. "What’s in an act? On security speech acts and little security nothings." Security Dialogue 42, no. 4-5 (2011): 371-383.

CE

UeT

DC

olle

ctio

n

33

desecuritizing process creates a structure of desecuritization that constrains actors’ ability to

convincingly speak security. Finally, in order to conceptualize how an institutional structure can be

seen as a possible desecuritization strategy, we must understand what the aim of desecuritization

should be: what is the nature of the politics it aims to restore?

Concerned with a “democratic deficit” in the CS theory of securitization, Aradau claims

desecuritization should be thought of as “the democratic challenge to the non-democratic politics of

securitization” and must “be inscribed institutionally…to create a different relation from the one of

enmity, a relation which is not rooted in the exclusionary logic of security” [emphasis added].76 She

advocates an emancipatory model of desecuritization, whereby those who are constructed as

threatening claim for themselves the already-established universal principles of the state, shedding

their securitized identity and embracing that of equal citizen.77 Roe and Jutila correctly dismiss this

strategy for desecuritizing national minorities, as it means the death of the distinctive minority

identity and thus the group itself. Another issue with the emancipation argument is that while

Aradau claims it is a strategy for restoring democratic politics, she implicitly views democracy as

majoritarian democracy.

2.5 What kind of politics?

What sort of democratic politics can desecuritization restore? Citing Michael Saward, Aradau

gives us one broad idea of what kind of democracy emancipation (re)establishes: 1) political equality

and fairness, 2) policy stemming from popular power, and 3) transparency and the possibility of

public scrutiny.78 She goes on to link emancipation with this broad notion of democracy, believing

that emancipation can disrupt “the exclusionary logic of security” and simultaneously establish “a

76 Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation.” 77 Ibid., 401-402. 78 Ibid., 392.

CE

UeT

DC

olle

ctio

n

34

new relationality with the other.”79 However, none of this can occur without dis-identification,

which as Behnke correctly recognizes, means, “the price of emancipation…is the elimination of

difference.”80 Consequently, Aradau’s image of democracy cannot be one of power-sharing or

consociation and cannot apply in a state divided along ethnic or national lines.

The kind of democracy that Aradau advocates might even be a facilitating condition for

(re)securitization in deeply divided states. For example, it is not difficult to imagine the following

scenario. In a country deeply divided along ethnic lines, the largest group attempts to redefine the

state as a centralized territory with a majoritarian political system, suppressing all forms of

nationalism and advocating a non-ethnic, civic identity. Subsequently, their language, customs and

culture are made the norm, while all others are suppressed. Referendums challenging any of these

norms are allowed in this hypothetical state, but they are easily defeated because the majoritarian

character of the state means the group that makes up just over 50% of the population can govern

alone. In such a country, it would be very difficult not to see a group securitize their identity vis-à-vis

the dominant group, or vice versa.

While this is an extreme example and no such state could be called a democracy, it illustrates

how majority rule can theoretically be simultaneously democratic and undemocratic. In practice,

divided states institute language rights, cultural rights etc. and may create incentives for elites to seek

support from other groups. Andrew Reynolds calls this approach integrative majoritarianism, –

associated with Horowitiz – where majoritarian structures prevail but contain centripetal forces that

encourage elites to be moderate and inclusive.81 Nonetheless, this theoretically reasonable idea is

79 Ibid., 401. 80 Behnke, Andreas. "No Way Out: Desecuritization, Emancipation and the Eternal Return of the Political — a Reply to Aradau," 67. 81 Reynolds, Andrew. "Majoritarian or Power-Sharing Government." In Democracy and Institutions: The Life Work of Arend Lijphart, Markus M.L Crepaz, Thomas A. Koelble, and David Wilsford, 155-196. Ann Arbor: University of Michigan Press, 2000, 156-160.

CE

UeT

DC

olle

ctio

n

35

near impossible to implement and most states today contain some form of power-sharing; purely

majoritarian states are quite rare.82

Aradau claims that desecuritization involves deciding what sort of politics we want, and her

answer, though uncertain, seems to preclude power-sharing democracy based on the necessity of

dis-identification. Inevitably, the kind of politics we want in a divided state are those that allow for

all the broad qualities associated with democracy – freedom, transparency, accountability, inclusivity,

and so on – but do not demand dis-identification along universalist principles. Consociational

democracy allows for all the individual freedoms commonly associated with democracy but more

importantly, it concedes that sometimes, individuals may feel the need to politicize a collective

identity. The key is to establish structures that allow politicization in order to prevent or end

securitization of ethnic identities. Thus, the kind of politics that desecuritization aims for must be a

form of democracy that is acceptable to all groups, while protecting basic individual liberties and the

rule of law.

The question of politics is crucial in theorizing strategies of desecuritization in deeply

divided, post-conflict states. It has been argued that some established desecuritization theories are

limited in that they view desecuritization as the restoration of liberal democratic politics. Such an

understanding of desecuritization may be valid in liberal democracies that have securitized migrant

identity, but cannot be helpful outside of this context. Consequently, desecuritization is equated with

democratization, meaning that without the latter, it is not possible to speak about desecuritization.

By returning to Waever’s three states of security, – security, insecurity, and a-security –, we see that

this conceptualization of desecuritization privileges only one end state of desecuritization: a state of

a-security. As Roe has correctly pointed out, however, desecuritization may also result in a state of

82 Lijphart, Arend. Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice, 2008, 125-36.

CE

UeT

DC

olle

ctio

n

36

security: the presence of a language of security but without a turn to emergency measures and

extraordinary politics.

In the following chapter, I analyze the political landscape of postwar Bosnia in order to

demonstrate how an institutional structure of desecuritization might effectively lead to a state of

security without escalation into emergency measures. Bosnia presents a case where elites interact in a

system where they are severely restricted in their ability to securitize ethnic identity. This is not so

much due to the presence of impediments to securitization – although these exist as well – but more

so because the highly decentralized state greatly empowers ethnic elites, giving them an incentive to

preserve the existing arrangements.

This decentralized system has been a double-edged sword: while ethnic identity remains

desecuritized and the possibility of renewed conflict is low, the country is crippled by nationalist

rhetoric. Nationalist politicians are generally unwilling to pursue reforms that would challenge their

privileged positions, meaning that passing the necessary reforms for EU accession has proven

extremely difficult. Nevertheless, it is important to remember that desecuritization need not aim at

restoring liberal democratic politics; in deeply divided, post-conflict states like Bosnia, any kind of

politics is better than no politics and violence.

CE

UeT

DC

olle

ctio

n

37

3. Institutional desecuritization in post-Dayton Bosnia

The war in Bosnia ended in 1995 with the signing of the Dayton Accords (Dayton) and the

establishment of highly decentralized state consisting of two entities, one predominately Serb and

the other split between Bosnian Muslims (Bosniaks) and Croats. Today, Bosnia is a case of moderate

securitization: ethnic identity is still a securitized issue, but it is dealt with in the sphere of normal

politics. Structural desecuritization was achieved through the implementation of consociational

democracy. While this has entrenched nationalist politics, the country is stable and democratizing at

a sluggish but steady pace.

How has Bosnia progressed from bloody inter-ethnic warfare to a stable, albeit still divided,

democracy in a relatively short period? Previous chapters have discussed various strategies of

desecuritizing ethnic identity, settling on a managing, institutional solution based on inter-ethnic

power sharing. This chapter will examine how Dayton established a political arrangement that

successfully desecuritized ethnic identity in Bosnia, creating the possibility of deliberation and

democratic. A procedural element was introduced that eased ethnic tensions to a manageable level.

Before turning to Bosnia’s institutional solution to securitized ethnic relations, a brief

evaluation of the conditions of securitization and subsequent conflict is warranted. After this, the

circumstances for desecuritization will be analyzed, elaborating specifically on the political situation

in Bosnia today. Specifically, I show how two instances of potential re-securitization were effectively

handled through the existing framework, that is, through normal politics. The purpose here is to

demonstrate that desecuritization does not need lofty goals of re-instating a liberal democracy: an

imperfect political community is better than no politics at all. Bosnia today is far from being a

vibrant, consolidated democracy, but ethnic identity has been desecuritized by consociational

arrangements, creating the possibility of further democratization that could not have been possible

right after the war.

CE

UeT

DC

olle

ctio

n

38

3.1 From insecurity to security: the success of the Dayton Accords

Upon witnessing the images of destruction, ethnic cleansing, and violence on a scale unseen

in Europe for over forty years, few people in the world believed Bosnia could rise from the ashes as

a democratic multi-ethnic state.83 Many still doubt whether it can ever again be the cosmopolitan,

tolerant “Switzerland of the Balkans”, which it was represented as during the 1984 Winter Olympics

in Sarajevo. Furthermore, it was doubtful that the Dayton treaty would actually bring lasting peace,

as all parties were pressured to sign it. Yet in the nearly twenty years since it was signed, peace has

prevailed and most importantly, few people see war as a way of solving political problems.

The main political issue that divided Bosnia in the early 1990s remains unresolved today: the

structure and identity of a Bosnian state. With the declaration of independence in early 1992, the

country became divided between Bosnian Serbs on one side and Bosnian Croats and Bosniaks on

the other. Most Serbs did not want to separate from Yugoslavia and boycotted the independence

referendum. Swept up by nationalist propaganda, Bosnian Serbs came to believe that an independent

Bosnia was a threat to their very identity. In Serb media outlets, Croats were portrayed as genocidal

fascists bent on exterminating Serbs, while the Bosniaks were represented as radical fundamentalists

wanting to re-establish Islamic rule over the Serbs.

It is unclear to what extent people believed these accounts, but what is clear is that many felt

existentially threatened as a people. In response to the declaration of independence, the Serb-

dominated Yugoslav National Army (JNA) intervened in Bosnia under the premise of protecting the

peace while in actuality supporting the secessionist policies of the newly declared Bosnian Serb

Republic on the territory of Bosnia. Bosnian Croats initially fought alongside the Bosniaks, but

subsequently established their own secessionist state of Herceg-Bosna with support from Franjo

83 See for example, Mearsheimer, John J., and Robert A. Pape. "The Answer: A Partition Plan for Bosnia." The New Republic, June 14, 1993

CE

UeT

DC

olle

ctio

n

39

Tudjman’s regime in neighboring Croatia.84 A comprehensive review of the war in Bosnia is

unnecessary, it is important only to demonstrate that Serbs and Croats saw an independent, unitary

Bosnia as a threat to their identity. The campaigns of ethnic cleansing against them led to Bosniaks

viewing their own identity as existentially threatened as well; they saw a unitary, multiethnic Bosnia

as the only guarantee of their survival as a people.

At the height of the war, survival for the Croats and the Serbs of Bosnia meant having their

own state or being part of Croatia and Serbia respectively, while Bosniaks wanted a strong

centralized state. Why did all sides then agree to a treaty where they would not only have to share a

decentralized state, but where the return of refugees – the reversal of all ethnic cleansing that had

occurred – was proclaimed to be one of the most important goals? How did the presence of other

ethnicities, arguably one of the causes of war, become an entirely acceptable outcome? While the

pressure put on the local leaders to reach a deal at Dayton was important, the lasting success of the

agreement has been the unique, decentralized consociational government that has made the re-

securitization of ethnic identity difficult. By institutionalizing ethnic politics at all levels of

government, the Dayton agreement has made convincing any group of the necessity of elevating

ethnic issues above politics extremely difficult. Before turning directly to the proposition that

structural desecuritization has proven to be a useful strategy in Bosnia, more must be said about the

political structure itself.

3.2 Consociational democracy in Bosnia

When Dayton was signed, it created what would become one of the world’s most

decentralized states. Decision-making takes place on four levels: state, canon, entity and municipal;

most of the power rests in the municipal governments and the entities, the latter of which are

84 For a review of the war, as well as its immediate causes, see Little, Alan, and Laura Silber. Yugoslavia: Death of a Nation. London: Penguin, 1997.

CE

UeT

DC

olle

ctio

n

40

charged with providing all necessary funds for the state level. The entities are perhaps the most

important level, as they broadly represent the wartime arrangements: the Serbs reside primarily in the

autonomous Republika Srpska (RS), while the Federation of Bosnia-Herzegovina (FBiH) is split

between Bosniaks and Croats. The electoral rules are complex and differ at each level, containing

both majoritarian and consociational elements. Finally, reserved seats and proportional

representation rules ensure that groups are represented even in places where they are not the

majority.85

Dayton created a state that was just centralized enough to be considered a singular state and

decentralized enough to ease Serb and Croat fears of subjugation under Bosniak majority rule. There

is both territorial and political autonomy, and power sharing is enshrined in every level of

government. Crucially, each group was given a veto rights over legislation they perceived to be

against their “vital national interests,” a move that has been critiqued for being necessary but poorly

implemented.86 The veto was implemented in order to guarantee that no law is passed without

consensus among the three groups, but the vagueness of “vital national interests” has led to elites

using the veto power in a wide array of cases.87

The ability for all groups to exercise veto power on almost any issue regarding ethnic

interests has been a curse and a blessing. In terms of democratization, it has meant that nationalist

politicians can be fully uncompromising, knowing full well that they can polarize their constituency

by demonstrating how another group wants to dominate them. Consequently, the pace of reform

towards EU accession has been painfully slow, as the use of the veto naturally leads to maintaining

85 Bieber, Florian. Post-War Bosnia Ethnicity, Inequality and Public Sector Governance. Houndmills: Palgrave Macmillan, 2006, 51-81. 86 Bieber, Florian, and Soeren Keil. "Power-Sharing Revisited - Lessons learned in the Balkans?" Review of Central and East European Law 34, no. 4 (2009): 337-360. 87 Bahtic-Kunrath, Birgit. "Of veto players and entity-voting: institutional gridlock in the Bosnian reform process." Nationalities Papers 39, no. 6 (2011): 899-923.

CE

UeT

DC

olle

ctio

n

41

the state quo.88 On the other hand, the possession of an institutionally guaranteed and vaguely

defined veto has meant that no law can be passed which would threaten the nationalist interests of

any group. Consequently, securitizing an issue is an unnecessary option, as the existing structure

makes it possible to deal with any existential threat through normal politics.

In spite of rampant nationalist rhetoric, political deadlock, and persistent ethnic divisions,

the ethnic issue in Bosnia remains desecuritized. The consociational framework created by Dayton

has drastically limited the ability of actors to securitize the ethnic issue. Breaking free from this

structure would not only be highly de-stabilizing, it would be extremely difficult to justify politically,

as ethnic identity is heavily protected through various mechanisms like the veto discussed above.

Furthermore, elites have strong incentives to work within the current structure rather than attempt a

securitizing move and lose legitimacy if it fails. This structurally produced desecuritization is most

visible in elites’ unwillingness to try to cross beyond the threshold of established rules, even in times

of political crisis.

3.3 Institutional desecuritization in times of crisis

Since 1995, Bosnia has faced numerous crises that have threatened peace and stability. Until

2004, the presence of 60,000 NATO troops did much to guarantee concord between the three

groups, ensuring that any crisis could effectively be contained before it spun out of control. Perhaps

the best example is when NATO troops stormed the offices of the Hrvatska Demokratska Zajednica’s

(HDZ) financial supporters and the OHR dismissed many of its leaders, ending a dispute that nearly

tore apart the Bosniak-Croat Federation.89 Since 2004, however, the Stabilization Force (SFOR) has

88 Ibid., 918. 89 Bose, Sumantra. Bosnia after Dayton: nationalist partition and international intervention. Cambridge: Oxford University Press, 2002.

CE

UeT

DC

olle

ctio

n

42

been replaced by a much smaller EU-led force (EUFOR) of only 1,200 troops stationed today.90 It

could be argued that the early political crises Bosnia faced were in large part mitigated by the sizeable

presence of foreign troops. Later crises, on the other hand, were averted by the structure of

desecuritization.

The period following the failure of constitutional reform in 2006 was marked by an increase

in political deadlock, rising nationalist rhetoric, and a general decline in Bosnia’s progress towards

democratization and EU accession.91 Possibility of armed conflict again seemed plausible, albeit

unlikely.92 In this period, two major crises related to the 2010 elections arose. First, the 2010

elections resulted in a strong victory for the Socijaldemokratska Partija (SDP), who formed a

government in the FBiH without the inclusion of the two biggest Croat parties, the HDZ and the

HDZ-1990. As a result, the two HDZ parties protested the government, calling it illegal because it

was formed without real representation for the Bosnian Croats (the SDP had formed a coalition

with minor Croat parties, who had less than 10% of the Croat vote). Bosnia’s central election

committee quickly ruled against the SDP, deeming the government illegal; more than a year passed

before Bosnia had a government, the sides unable to compromise on a workable coalition.93

A second crisis arose shortly after the 2010 elections when Milorad Dodik, President of the

RS, suggested that he might call for a referendum on Bosnia’s state-level judiciary. Citing the fact

that the state courts prosecute Serbs more than others, Dodik claimed that the courts were biased

and not working in the interest of the Bosnian Serbs.94 Such a referendum would have called into

question the authority of the OHR and the legitimacy of the state itself, while simultaneously setting

90 EUFOR. "EUFOR Fact Sheet." Accessed May 27, 2012. http://www.euforbih.org/index.php?option=com_content&view=article&id=15&Itemid=134. 91 Chivvis, Christopher S., and Harun Dogo. "Getting Back on Track in Bosnia-Herzegovina." The Washington Quarterly 33, no. 4 (2010): 104-109. 92 Chivvis, Christopher S. "Back to the Brink on Bosnia?" Survival: Global Politics and Strategy 52, no. 1 (2010): 97-110. 93 Karic, Mirsad. "Bosnian General Elections of 2010 and the Post-Election Crisis." Bilge Strateji 4, no. 6 (2012): 71-103. 94 Hadzovic, Eldin, and Drazen Remikovic. "Bosnia: Dodik Agrees to Drop Disputed Referendum." Balkan Insight, May 11, 2011. Accessed May 29, 2012. http://www.balkaninsight.com/en/article/catherine-ashton-to-meet-bosnian-leaders.

CE

UeT

DC

olle

ctio

n

43

a precedent for the RS to hold a referendum on other issues as well, mainly secession from Bosnia.95

In the end, Dodik backed down from calling for the referendum, and the SDP was able to form a

government with the other main ethnic parties. Unquestionably, international pressure helped in

deflecting both crises; but the willingness of elites to work within the existing structure – rather than

to step outside of it – was instrumental in preventing securitization over these very important issues.

Although he makes his contempt for Bosnia very clear and often suggests the possibility of

its dissolution, no politician in Bosnia today benefits more from the Dayton framework than

Milorad Dodik. Independence for the RS, though a popular idea, is not in Dodik’s interest. His

influence and authority is greatest with the current arrangement, where he has greatly consolidated

his hold on power and demands the attention of top-level European diplomats like Catherine

Ashton, the EU’s High Representative for Foreign Affairs and Security Policy.96 In effect, Dayton

has given the Bosnian Serb leadership what three years of warfare could never achieve: autonomy

and international legitimacy.

Dodik may express displeasure with the central government and hint at the possibility of

secession, but as the past six years have demonstrated, he is unwilling to go the final steps in

challenging the established framework. Bosnian Serbs are quite happy with the current autonomy

given to them by Dayton; their main qualm is with Bosniak politicians who wish to do away with the

entities and institute a majoritarian, centralized state. For example, in the 2006 elections, Haris

Silajdzic of the Stranka za Bosnu i Hercegovinu (SBiH), a Bosniak nationalist party, ran on a platform of

eliminating the entities and centralizing Bosnia. In turn, Dodik was able to gain support by pointing

to Silajdzic’s popularity as evidence that Bosniaks threatened Bosnian Serb autonomy.

95 Parish, Matthew. "Bosnia: Dodik Agrees to Drop Disputed Referendum." Balkan Insight, May 16, 2011. Accessed May 29, 2012. http://www.transconflict.com/2011/05/milorad-dodik-and-the-politics-of-referendum-165/. 96 Parish, Matthew. "Bosnia: Dodik Agrees to Drop Disputed Referendum."

CE

UeT

DC

olle

ctio

n

44

While the consociational arrangement does generate centrifugal nationalist politics, it also

makes securitization difficult and undesirable from a political standpoint. The high level of

decentralization afforded to the RS, coupled with the veto on national interests, represent a

structural desecuritization that protects Bosnian Serb interests while simultaneously making

securitization less likely. A shift in either direction – whether it is greater centralization or outright

independence – would destroy this balance and severely hurt Bosnian Serb interests. Significant

centralization and the elimination of their entity could lead to them being outvoted and dominated

by more numerous Bosniaks, while independence would leave them in a position similar to

Transnistria, only with much less viable borders and no significant Russian support. If Bosniaks

pushed for centralization and the elimination of entities, however, re-securitization and war could

ensue; but the drastic fall in Silajdzic’s popularity and the success of the moderate SDP suggests this

will not happen in the near future.97

Bosnia’s decentralization and the creation of the RS ensures that Bosnian Serb politicians

will not try to securitize identity issues, as doing so would risk endangering their interests. Croats in

Bosnia are in a much more precarious situation, as they share the FBiH with the Bosniaks, who

greatly outnumber them. The Croat representative at the state level presidency, Zeljko Komsic, was

elected due to large numbers of Bosniaks voting for him, meaning he is not viewed as a legitimate

representative of Croat interests by the Croatian nationalist parties.98 Frustration with the current

arrangement was exacerbated in the 2010 elections when the SDP formed a coalition government

without the two main Croat nationalist parties, effectively excluding a vast majority of the Croat

electorate from power. The government coalition was ruled to be in violation of basic power sharing

97 Toal, Gerard, and Adis Maksic. "Is Bosnia-Herzegovina Unsustainable? Implications for the Balkans and European Union."Eurasian Geography and Economics 52, no. 2 (2011): 285-286. 98 Karic, Mirsad. "Bosnian General Elections of 2010 and the Post-Election Crisis,” 80-81.

CE

UeT

DC

olle

ctio

n

45

principles and the Croat nationalist parties condemned their continued exclusion from the center of

power, arguing for the necessity of a third Croat entity.99

The political crisis following the 2010 elections had all the right ingredients for a re-

securitization of ethnicity in Bosnia and the renewal of armed conflict. Bosnian Serbs under Dodik

were fearful as ever of any attempts to dismantle the RS, while Bosnian Croats saw their share of

power greatly diminished in the FBiH by the SDP’s unwillingness to include their main parties in its

coalition. Furthermore, the international community, suffering its own turmoil due to the economic

crisis, seemed unable to convince the leaders to work together. Yet the SDP did eventually form a

coalition with the Croat parties (after more than a year of negotiating), and the relatively quick

adoption of a new state budget in spring of 2012 suggests that this coalition might even be able to

work together to pass necessary reforms.

Why did Bosnia not fall back into warfare and insecurity following the 2010 election crisis?

As always, there is more than one possible answer. One important factor is that unlike in the early

1990s, Croatia and Serbia are much less willing to support secessionism in Bosnia, being

preoccupied with their own domestic issues. Additionally, even though the EU was struggling with

the financial crisis, it is unlikely they would have tolerated the sort of instability that armed conflict

brings. Nonetheless, these alone cannot explain the relatively successful resolution of Bosnia’s crisis.

In an exemplary consociational manner, Bosnia’s elites compromised on crucial issues when

they realized no one would step in to solve their problems. Though far from perfect, the

consociational framework provides a basic structure for leaders to settle identity-related political

issues without resorting the securitization. Political participation and autonomy are crucial for

sustained peace and continuance of procedural politics, and Dayton ensures both of these for all

three dominant communities. Nevertheless, it is a fundamentally flawed system that cannot survive

99 Touquet, Heleen. "Multi-Ethnic Parties in Bosnia-Herzegovina: Nasa Stranka and the Paradox of Postethnic Politics." Studies in Ethnicity and Nationalism 11, no. 3 (2011): 460.

CE

UeT

DC

olle

ctio

n

46

in the long run. For the time being, however, it is better that ethnic divisions are managed in a

flawed but peaceful and deliberative way rather than in no way at all. Regrettably, the choice for

Bosnia today is not between liberal democracy and nationalist politics, but between nationalist

politics or no politics at all.

In many ways, Dayton’s success in desecuritization stems from the highly decentralized state

it created. A decentralized state was the only way Serbs and Croats would accept unity with the

Bosniaks; guarantees of extensive autonomy eased their fears of being dominated. Furthermore,

even a highly decentralized state like the one proposed in Dayton is still a state, with internationally

recognized borders and a political structure that encompasses one defined community. Critics note

that the effectiveness of the central government is very weak and that real power lies in the

homogenous entities and municipalities. This is a valid point, but it fails to appreciate the bigger

picture: the question in Bosnia is no longer whether Bosnia should exist but what sort of democracy

it should be – a far better position than it was in in 1995. Desecuritization does not necessarily mean

democratization, although any desecuritizing move should not prevent the growth of genuine

democracy in the future.

I have provided an account of how institutional structures shape actors’ ability and

willingness to securitize ethnic identity in Bosnia. Taking into account the high level of

decentralization, protection of ethnic interests through mutual vetoes, and some nationalist

politicians’ happiness with the status quo, it has been suggested that desecuritization has been

achieved institutionally. Furthermore, the persistence of nationalist politics and discussion of ethnic

issues does not preclude a state of desecuritization: Bosnia’s Dayton framework, while far from

perfect, has indeed desecuritized the ethnic issue because even in times of crisis, ethnic leaders

choose to work within the constitutional framework rather than escape it.

CE

UeT

DC

olle

ctio

n

47

Crucially, this does not mean that the Dayton framework allows for a smooth transition

towards a functioning democracy. Rather, the current political landscape should be viewed as a

necessary transitional step, smoothing over relations, easing tensions, and creating the basis for a

common political community that can solve its problems through established rules and norms

instead of with violence. Critics of the Dayton system are correct in pointing out its many flaws that

hinder democratization and European integration, but perhaps they forget that the war that

introduced the term ethnic cleansing into popular discourse is still fresh, even in the minds of young

adults. A deeply flawed arrangement, Dayton nevertheless remains the only thing keeping Bosnia

together and on the path towards Europe.

CE

UeT

DC

olle

ctio

n

48

Conclusions In reviewing four leading strategies of desecuritization, this paper has challenged

conventional ways of thinking about desecuritization. Particularly, it has suggested that deeply

divided, post-conflict societies offer unique challenges to conceptualizing desecuritization strategies.

Unlike the liberal democratic context in which much of the existing literature is based in, deeply

divided states have very weak democratic institutions or none at all. Consequently, desecuritization

in these states cannot mean the restoration of liberal democratic politics, as some have suggested,

but rather initiation of any political arrangement that manages to successfully bring the securitized

issue out of the realm of emergency measures. In other words, while desecuritization in deeply

divided, post-conflict states might not be able to introduce particularly robust democratic politics, its

merit should not be judged based on this alone.

This requires conceptualizing a different security arrangement than has been embraced by

much of the existing literature on desecuritization: rather than bringing about a-security,

desecuritization can also lead to a state of security. If a-security means that the language of security

and emergency politics are both non-existent, a state of security indicates that while there is still a

discourse of security surrounding a particular issue, it is nevertheless dealt with within the realm of

normal politics. In post-conflict states, it might very be the case that society must first go through a

transitional state of security before the language of security withers away completely. This was the

way in which the European non-war community emerged in the latter half of the 20th century and it

might well be the case for many deeply divided, post-conflict states today.

Analyzing how structure and institutions play a role in elevating political crises in Bosnia, this

paper has argued that desecuritization can be institutionalized, altering the facilitating conditions for

actors to speak security. The possession of mutual vetoes at most levels of government combined

with a state structure that benefits entrenched elites, Bosnia is a typical case of a consociational

CE

UeT

DC

olle

ctio

n

49

arrangement, where elites work together to stabilize the system. This arrangement is nevertheless

deeply flawed, as the necessary reforms for EU accession are extremely difficult to implement,

impeding the country’s European integration. On the other hand, if one considers the position

Bosnia was in a mere two decades ago, it is clear that consociational power sharing may have am

important role in aiding deeply divided, post-conflict states’ transitions from insecurity to a-security

regarding ethnic issues.

What are some points of departure for further research? First, it is clear that the CS’s

theories have some limitations outside of a liberal democratic context. In particular, because of the

preference for desecuritization, new theories should be constructed to deal with different paradigms

the CS has not yet approached. Second, while this paper has suggested that power sharing or

consociation might be a useful desecuritization strategy in deeply divided, post-conflict states, it has

only looked at one case: Bosnia. One interesting case to consider might be that of Northern Ireland,

where inter-communal tensions seem to have greatly faded with the signing of the Good Friday

Agreement and the introduction of power sharing mechanisms. The lessons learned from these

explorations could prove to be indispensable in state-building endeavors in the future.

Finally, in developing desecuritization strategies, the field of security studies would do well to

consult their colleagues in other fields, who have been confronting similar challenges for decades.

Desecuritization strategies comprise only a very small amount of literature in the field of security

studies: expanding this literature is a worthwhile endeavor not only for developing a more robust

understanding of what security is in international relations, but for the pursuit of peace as well.

CE

UeT

DC

olle

ctio

n

50

Works Cited Aradau, Claudia. "Security and the Democratic Scene: Desecuritization and Emancipation." Journal of International Relations and Development 7, no. 4 (2004): 388-413. Bahtic-Kunrath, Birgit. "Of veto players and entity-voting: institutional gridlock in the Bosnian reform process." Nationalities Papers 39, no. 6 (2011): 899-923. Balibar, Étienne. "The Nation Form: History and Ideology." In Becoming National: A Reader., Geoff Eley and Ronald G. Suny, New York: Oxford University Press, 1996. Behnke, Andreas. "No Way Out: Desecuritization, Emancipation and the Eternal Return of the Political — a Reply to Aradau."Journal of International Relations and Development 9, no. 1 (2006): 62-69. Bieber, Florian, and Soeren Keil. "Power-Sharing Revisited - Lessons learned in the Balkans?" Review of Central and East European Law 34, no. 4 (2009): 337-360. Bieber, Florian. Post-War Bosnia Ethnicity, Inequality and Public Sector Governance. Houndmills: Palgrave Macmillan, 2006. Bigo, Dider. "Security and Immigration: Toward a Critique of the Governmentality of Unease." Alternatives: Global, Local, Political 27, no. 1 (2002): 63-92. Bose, Sumantra. Bosnia after Dayton: nationalist partition and international intervention. Cambridge: Oxford University Press, 2002. Brubaker, Rogers. Nationalism Reframed: Nationhood and the National Question in the New Europe. Cambridge: Cambridge University Press, 1996. Buzan, Barry, and Lene Hansen. The evolution of international security studies. Cambridge: Cambridge University Press, 2009. Buzan, Barry, Ole Waever, and Jaap de Wilde. Security: a new framework for analysis. Boulder: Lynne Rienner, 1998. Buzan, Barry, and Ole Waever. "Slippery? contradictory? sociologically untenable? The Copenhagen schools replies." Review of International Studies 23 (1997): 241-250. Ceyhan, Ayşe, and Anastassia Tsoukala. "The Securitization of Migration in Western Societies: Ambivalent Discourses and Policies." Alternatives: Global, Local, Political 27 (2002): 21-39. Chivvis, Christopher S., and Harun Dogo. "Getting Back on Track in Bosnia-Herzegovina." The Washington Quarterly 33, no. 4 (2010): 103-118. Chivvis, Christopher S. "Back to the Brink on Bosnia?" Survival: Global Politics and Strategy 52, no. 1 (2010): 97-110.

CE

UeT

DC

olle

ctio

n

51

EUFOR. "EUFOR Fact Sheet." Accessed May 27, 2012. http://www.euforbih.org/index.php?option=com_content&view=article&id=15&Itemid=134. Geertz, Clifford. The Interpretation Of Cultures. New York: Basic Books, 1973. Guzzini, Stefano. "Securitization as a causal mechanism." Security Dialogue 42, no. 4-5 (2011): 329-341. Hadzovic, Eldin, and Drazen Remikovic. "Bosnia: Dodik Agrees to Drop Disputed Referendum." Balkan Insight, May 11, 2011. Accessed May 29, 2012. http://www.balkaninsight.com/en/article/catherine-ashton-to-meet-bosnian-leaders. Hansen, Lene. "The Little Mermaid's Silent Security Dilemma and the Absence of Gender in the Copenhagen School."Millennium - Journal of International Studies 29, no. 2 (2000): 285-306. Hobsbawm, E J. Nations and Nationalism Since 1780: Programme, Myth, Reality. 2nd ed. Cambridge: Cambridge University Press, 1992. Horowitz, Donald L. Ethnic Groups in Conflict. Berkeley: University of California Press, 1985. Huysmans, Jef. "What’s in an act? On security speech acts and little security nothings." Security Dialogue 42, no. 4-5 (2011): 371-383. Huysmans, Jef. "The European Union and the Securitization of Migration." Journal of Common Market Studies 38, no. 5 (2000): 751-777. Huysmans, Jef. "The Question of the Limit: Desecuritisation and the Aesthetics of Horror in Political Realism." Millennium - Journal of International Studies 27, no. 3 (1998): 569-589. Huysmans, Jef. "Migrants as a Security Problem: Dangers of "Securitizing” Societal Issues." In Migration and European Integration. The Dynamics of Inclusion and Exclusion, Dietrich Thränhardt and Robert Miles, 53-72. London: Pinter, 1995. Jutila, Matti. "Desecuritizing Minority Rights: Against Determinism." Security Dialogue 37, no. 2 (2006): 167-185. Karic, Mirsad. "Bosnian General Elections of 2010 and the Post-Election Crisis." Bilge Strateji 4, no. 6 (2012): 71-103. Kinnvall, Catarina. "Globalization and Religious Nationalism: Self, Identity, and the Search for Ontological Security." Political Psychology 25, no. 5 (2004): 741-767. Kymlicka, Will. "Justice and security in the accommodation of minority nationalism." In Ethnicity, Nationalism, and Minority Rights, Stephen May, Tariq Modood, and Judith Squires, 144-174. Cambridge: Cambridge University Press, 2004. McDonald, Matt. "Securitization and the Construction of Security." European Journal of International Relations 14, no. 4 (2008): 563-587.

CE

UeT

DC

olle

ctio

n

52

Lijphart, Arend. Thinking about Democracy: Power Sharing and Majority Rule in Theory and Practice. London: Routledge, 2008. Lijphart, Arend. The Politics of Accommodation: Pluralism and Democracy in the Netherlands. Berkeley: University of California Press, 1968. Little, Alan, and Laura Silber. Yugoslavia: Death of a Nation. London: Penguin, 1997. Mearsheimer, John J., and Robert A. Pape. "The Answer: A Partition Plan for Bosnia." The New Republic, June 14, 1993 McSweeney, Bill. "Identity and Security: Buzan and the Copenhagen School." Review of International Studies 22, no. 1 (1996): 81-93. Micgiel, John S. State and Nation Building in East Central Europe: Contemporary Perspectives. New York: Columbia University Press, 1996. Mitzen, Jennifer. "Ontological Security in World Politics: State Identity and the Security Dilemma." European Journal of International Relations 12, no. 2 (2006): 341-370. Norris, Pippa. Driving Democracy: Do Power-Sharing Institutions Work? Cambridge: Cambridge University Press, 2008. O'Leary, Brendan. "Debating Consociational Politics: Normative and Explanatory Arguments." In From Power Sharing to Democracy Post-Conflict Institutions in Ethnically Divided Societies, Sid Noel, Toronto: McGill-Queens University Press, 2005. Parish, Matthew. "Bosnia: Dodik Agrees to Drop Disputed Referendum." Balkan Insight, May 16, 2011. Accessed May 29, 2012. http://www.transconflict.com/2011/05/milorad-dodik-and-the-politics-of-referendum-165/. Reynolds, Andrew. "Majoritarian or Power-Sharing Government." In Democracy and Institutions: The Life Work of Arend Lijphart, Markus M.L Crepaz, Thomas A. Koelble, and David Wilsford, 155-196. Ann Arbor: University of Michigan Press, 2000. Roe, Paul. "Reconstructing Identities or Managing Minorities? Desecuritizing Minority Rights: A Response to Jutila." Security Dialogue 37, no. 3 (2006): 425-438. Roe, Paul. "Securitization and Minority Rights: Conditions of Desecuritization." Security Dialogue 35, no. 3 (2004): 279-294. Sisk, Timothy D. Power Sharing and International Mediation in Ethnic Conflicts. Washington DC: United States Institute of Peace, 1996. Toal, Gerard, and Adis Maksic. "Is Bosnia-Herzegovina Unsustainable? Implications for the Balkans and European Union."Eurasian Geography and Economics 52, no. 2 (2011): 279-293.

CE

UeT

DC

olle

ctio

n

53

Touquet, Heleen. "Multi-Ethnic Parties in Bosnia-Herzegovina: Nasa Stranka and the Paradox of Postethnic Politics." Studies in Ethnicity and Nationalism 11, no. 3 (2011): 451-467. Waever, Ole. "The EU as a Security Actor: Reflections from a Pessimistic Constructivist on Post-Sovereign Security Orders." In International Relations Theory and the Politics of European Integration: Power, Security, and Community, Morten Kelstrup and Michael C. Williams, 250-294. London: Routledge, 2000. Waever, Ole. "Insecurity, security, and asecurity in the West European non-war community." In Security Communities, Emanuel Adler and Michael Barnett, 69-118. Cambridge: Cambridge University Press, 1998. Wilkinson, Claire. "The Copenhagen School on Tour in Kyrgyzstan: Is Securitization Theory Useable Outside Europe?"Security Dialogue 38, no. 1 (2007): 5-25.